SERVING THE QUANTITATIVE FINANCE COMMUNITY

 
User avatar
DoubleTrouble
Topic Author
Posts: 83
Joined: November 9th, 2010, 9:46 pm

Topic suggestion for P.hD. in mathematics

January 25th, 2013, 2:47 pm

Hi,I recently finished my M.Sc. in mathematics with finance as my specialization. Although I'd probably love a job in the industry, I have a "need" to learn more theoretical knowledge. I'm thinking of applying for a Ph.d-position in mathematics and specialize even more in mathematical finance. And if I get tired of the academic life after 5 years of further studies I believe that it will be easier to get a good job in the industry as a Ph.d. Even though most institutions do not require it, I think that it is qualifying to display an interest in a research field. The problem is that I find everything to be interesting. Some of the fields I have touched upon in my education and thesis are Stochastic Calculus (Ito calculus, stochastic partial differential equations, Malliavin calculus), Numerical Methods in finance (FD, FEM, etc.), and Pricing and hedging (stochastic vol, calibration, market imperfections).I need some guidance, what can I do to up my chances of getting a Ph.d-position?
Last edited by DoubleTrouble on January 26th, 2013, 11:00 pm, edited 1 time in total.
 
User avatar
quantstart
Posts: 154
Joined: March 7th, 2010, 11:16 am

Topic suggestion for P.hD. in mathematics

January 25th, 2013, 7:49 pm

Hi DoubleTrouble,I would first ask yourself whether you want to have a career in academia subsequent to your PhD. Will you be happy on the "post-doc train" until you become assistant professor? 5 years is a significant investment of your life and you will need to factor in your opportunity cost. It does sound that you are very enthused about your subjects (SDEs, FDM, FEM, Pricing), but bear in mind that the market for financial engineers is not what it used to be, even a few years ago. Many of today's new quants are now either well-paid developers ("quant devs"), head into risk management or are being hired by funds to become algo researchers. If you do end up heading down the financial engineer route, make sure that you enjoy both the mathematical algorithm development as well as the programmatic implementation, as you will be spending a substantial portion of your time implementing models (C++, C#/Java, Python, MatLab, R etc). Targeting investment banks via recruiters subsequent to your PhD is your best bet.As for upping your chances of getting a PhD position, the simple answer is to make sure you have extremely solid mathematics grades and as many top letters of recommendations as you can get from your faculty. The latter should not be underestimated, as it is a key differentiator between top students when competing for PhD positions.Have you decided on any particular schools you might head to?
 
User avatar
Cuchulainn
Posts: 63379
Joined: July 16th, 2004, 7:38 am
Location: Amsterdam
Contact:

Topic suggestion for P.hD. in mathematics

January 26th, 2013, 2:35 pm

QuoteOriginally posted by: DoubleTroubleHi,I recently finished my M.Sc. in mathematics with finance as my specialization. Although I'd probably love a job in the industry, I have a "need" to learn more theoretical knowledge. I'm thinking of applying for a Ph.d-position in mathematics and specialize even more in mathematical finance. And if I get tired of the academic life after 5 years of further studies I believe that it will be easier to get a good job in the industry as a Ph.d. Even though most institutions do not require it, I think that it is qualifying to display an interest in a research field. The problem is that I find everything to be interesting. Some of the fields I have touched upon in my education and thesis are Stochastic Calculus (Ito calculus, stochastic partial differential equations, Malliavin calculus), Numerical Methods in finance (FD, FEM, etc.), and Pricing and hedging (stochastic vol, calibration, market imperfections).I need some guidance, what can I do to up my chances of getting a Ph.d-position?For a PhD (assuming you have hard maths under your belt) you probably need to be _very_ focused, e.g. only one of FEM(?), FDM or SDE. Having chosen then focus on class of DE. A good knowledge of functional and numerical analysis is almost mandatory unless you're happy applying 'recipes'.
Last edited by Cuchulainn on January 25th, 2013, 11:00 pm, edited 1 time in total.
My C++ Boost code gives
262537412640768743.999999999999250072597198185688879353856337336990862707537410378210647910118607313

http://www.datasimfinancial.com
http://www.datasim.nl
 
User avatar
ppauper
Posts: 70239
Joined: November 15th, 2001, 1:29 pm

Topic suggestion for P.hD. in mathematics

January 27th, 2013, 7:37 am

are you intending to stay at the school where you did your MSc ? working with people is important for a PhD, and you should be asking the profs you like at your school for possible projects.You could be the smartest guy in the world but if there's no one to supervise what you want to do, they won't take you
 
User avatar
DominicConnor
Posts: 11684
Joined: July 14th, 2002, 3:00 am

Topic suggestion for P.hD. in mathematics

January 27th, 2013, 7:00 pm

Cuchulainn is right that FDM,FEM and of course MC should be on your shopping list of skills to acquire.What makes it more complex is that numerical methods are an old discipline, one technique I was taught predates Arabic/decimal notation ( some kudos for the person to work out which)Given that a PhD is research degree, your task will be to get to the edge of the field and extend it and you will find that most of the numerical methods in use data from before computers.I'm glad that you find this stuff interesting since I routinely put the frighteners on people who think a PhD should be determined by career aspirations, but...Whether your future is in academia or banking, I'd like you to consider which areas you believe yourself to be better than other people ?
 
User avatar
Cuchulainn
Posts: 63379
Joined: July 16th, 2004, 7:38 am
Location: Amsterdam
Contact:

Topic suggestion for P.hD. in mathematics

January 27th, 2013, 7:26 pm

And underlying these high-level F*M methods are the fundamental numerical methods that can save you when recipes break down (and they always do, sooner or later). And they can be applied outside the (FDM, FEM, MC) triad.
Last edited by Cuchulainn on January 26th, 2013, 11:00 pm, edited 1 time in total.
My C++ Boost code gives
262537412640768743.999999999999250072597198185688879353856337336990862707537410378210647910118607313

http://www.datasimfinancial.com
http://www.datasim.nl
 
User avatar
DoubleTrouble
Topic Author
Posts: 83
Joined: November 9th, 2010, 9:46 pm

Topic suggestion for P.hD. in mathematics

January 27th, 2013, 9:55 pm

QuoteOriginally posted by: quantstartI would first ask yourself whether you want to have a career in academia subsequent to your PhD. Will you be happy on the "post-doc train" until you become assistant professor? 5 years is a significant investment of your life and you will need to factor in your opportunity cost. Not to be rude or anything, but does it really matter? When i browse jobs here at Wilmott a huge amount of them requires a Ph.d in mathematics, mathematical statistics, or computer science. During the autumn I got to the final round interviews for a quant research position at a major Nordic bank. I spoke with the head of trading models several different occasions. The questions were on a very sophisticated level (requiring e.g. Mallivain calculus). Of course he was a Ph.d in math from Oxford, and he said that it is very hard to get a good quant job in UK or France if you are not a Ph.d. QuoteOriginally posted by: quantstartHave you decided on any particular schools you might head to?At the moment my main choice is Imperial College in London. I have good knowledge about some of the professors at Imperial, since I cited their work in my thesis, and read many of their articles. However, I realize that the competition is extremely high for a position there. My best chance is at my current university since I know every professor, and one of my teachers in functional analysis (i did very very well in that exam ) is the committee for choosing new ph.d. students. Also my thesis supervisor is very well-reputed professor and I hope he has good things to say about me. At least my grade was good, but he is very very reserved with praise (basically, everyone who is not Hörmander is bad. Even himself). Also, what is promising is that they are hiring a new professor in mathematical finance and numerical methods.
 
User avatar
QuantCentral
Posts: 45
Joined: January 24th, 2013, 9:16 pm

Topic suggestion for P.hD. in mathematics

January 28th, 2013, 7:32 am

If you plan to become a quant in the industry, you will be best prepared by working on numerical methods during your PhD studies. However, a PhD thesis on numerical methods does not imply there is no theory in it. Very often, you propose a new algorithm, implement it, get some nice empirical results, but then you cannot prove the convergence or convergence rate theoretically and you might have to change your research topic. Take a look at the Journal of Computational Finance. You might find what you want there.
 
User avatar
quantstart
Posts: 154
Joined: March 7th, 2010, 11:16 am

Topic suggestion for P.hD. in mathematics

January 28th, 2013, 8:51 am

QuoteOriginally posted by: DoubleTroubleNot to be rude or anything, but does it really matter? When i browse jobs here at Wilmott a huge amount of them requires a Ph.d in mathematics, mathematical statistics, or computer science. During the autumn I got to the final round interviews for a quant research position at a major Nordic bank. I spoke with the head of trading models several different occasions. The questions were on a very sophisticated level (requiring e.g. Mallivain calculus). Of course he was a Ph.d in math from Oxford, and he said that it is very hard to get a good quant job in UK or France if you are not a Ph.d.Indeed, fundamentally it comes down to maximising some metric of your choosing. That could be overall bank balance or it could be enjoyment with the work. For full disclosure, I did a PhD and enjoyed it immensely. Given what you've mentioned, I would state that you will thoroughly enjoy it - and as you said it is (almost) a requirement for a deep quantitative position.QuoteOriginally posted by: DoubleTroubleAt the moment my main choice is Imperial College in London. I have good knowledge about some of the professors at Imperial, since I cited their work in my thesis, and read many of their articles. However, I realize that the competition is extremely high for a position there. My best chance is at my current university since I know every professor, and one of my teachers in functional analysis (i did very very well in that exam ) is the committee for choosing new ph.d. students. Also my thesis supervisor is very well-reputed professor and I hope he has good things to say about me. At least my grade was good, but he is very very reserved with praise (basically, everyone who is not Hörmander is bad. Even himself). Also, what is promising is that they are hiring a new professor in mathematical finance and numerical methods.Again, for full disclosure, I went to Imperial (although not for financial mathematics, for fluid dynamics) and had a great time there. If you can get a place there I would certainly take it. One of my friends is currently studying the MSc in Finance and Risk Management. He is having a great time, besides having to get back into the groove of being a student at 30...!
 
User avatar
DoubleTrouble
Topic Author
Posts: 83
Joined: November 9th, 2010, 9:46 pm

Topic suggestion for P.hD. in mathematics

February 10th, 2013, 4:33 pm

QuoteOriginally posted by: quantstartQuoteOriginally posted by: DoubleTroubleAt the moment my main choice is Imperial College in London. I have good knowledge about some of the professors at Imperial, since I cited their work in my thesis, and read many of their articles. However, I realize that the competition is extremely high for a position there. My best chance is at my current university since I know every professor, and one of my teachers in functional analysis (i did very very well in that exam ) is the committee for choosing new ph.d. students. Also my thesis supervisor is very well-reputed professor and I hope he has good things to say about me. At least my grade was good, but he is very very reserved with praise (basically, everyone who is not Hörmander is bad. Even himself). Also, what is promising is that they are hiring a new professor in mathematical finance and numerical methods.Again, for full disclosure, I went to Imperial (although not for financial mathematics, for fluid dynamics) and had a great time there. If you can get a place there I would certainly take it. One of my friends is currently studying the MSc in Finance and Risk Management. He is having a great time, besides having to get back into the groove of being a student at 30...!Thank you very much for your answer. I really appreciate it. May I ask some more specific questions about how it is to take you Ph.d in London? How are the environment, working conditions, social benefits, salary etc.? The reason I ask is because I have discussed the idea of pursuing a Ph.d with 3 professors (whom I know well) at my school. Two of them took their Ph.d at Princeton and the other one from MIT. They all say that it was a really bad research environment for Ph.d students. You didn't get your own office, no respect or encouragement from senior researcher. You were basically just a student on scholarship. However, with that said there schools are brilliant in many other ways. They all said that Sweden, Denmark, or Germany is probably the best place in the world to take your Ph.d. from that perspective. It is more like a full time job, with decent salary (no need for scholarships), and all the social benefits that come with a job. I am very much aware that this was some time ago, and that things probably changed since then. I am really interested in hearing about your experience from London.Right now I am sitting here writing a body for my application letter. To my own university I don't need to be extremely formal (in fact, i think they can be putt off by it). Since there are many people that will read the letter I can't address it to a specific person. Usually I write "dead sir or madam" but I feel that it not appropriate to my own school. What is a more neutral way of starting off a application letter?Thank you in advance!
 
User avatar
Cuchulainn
Posts: 63379
Joined: July 16th, 2004, 7:38 am
Location: Amsterdam
Contact:

Topic suggestion for P.hD. in mathematics

February 10th, 2013, 4:48 pm

I think it is better to write 'dear sir or madam' than text in previous post if you want some response from the persons addressed.
Last edited by Cuchulainn on February 9th, 2013, 11:00 pm, edited 1 time in total.
My C++ Boost code gives
262537412640768743.999999999999250072597198185688879353856337336990862707537410378210647910118607313

http://www.datasimfinancial.com
http://www.datasim.nl
 
User avatar
Alan
Posts: 10468
Joined: December 19th, 2001, 4:01 am
Location: California
Contact:

Topic suggestion for P.hD. in mathematics

February 10th, 2013, 5:57 pm

Dear Committee Members (?)
 
User avatar
DoubleTrouble
Topic Author
Posts: 83
Joined: November 9th, 2010, 9:46 pm

Topic suggestion for P.hD. in mathematics

February 11th, 2013, 10:35 pm

Thank you Cuchulainn and Alan for your suggestions. I will finish my Application letter and in the end decide upon the greeting phrase.
 
User avatar
ppauper
Posts: 70239
Joined: November 15th, 2001, 1:29 pm

Topic suggestion for P.hD. in mathematics

February 12th, 2013, 9:24 am

QuoteOriginally posted by: CuchulainnI think it is better to write 'dear sir or madam' than text in previous post if you want some response from the persons addressed.indeed, they might take it as a death threat
 
User avatar
ppauper
Posts: 70239
Joined: November 15th, 2001, 1:29 pm

Topic suggestion for P.hD. in mathematics

February 12th, 2013, 9:26 am

QuoteOriginally posted by: DoubleTroubleThank you Cuchulainn and Alan for your suggestions. I will finish my Application letter and in the end decide upon the greeting phrase.that's good.More general would be "To whom it may concern"
ABOUT WILMOTT

PW by JB

Wilmott.com has been "Serving the Quantitative Finance Community" since 2001. Continued...


Twitter LinkedIn Instagram

JOBS BOARD

JOBS BOARD

Looking for a quant job, risk, algo trading,...? Browse jobs here...


GZIP: On